New Regression Learner App in R2017a
MATLAB® enables domain experts (engineers and scientists) to do data science. The Regression Learning app makes it easy to get started fitting regression models—a common task in machine learning. The app also makes it easy to:
- Get started with regression machine learning problems
- Quickly compare models and choose the best one
- Generate MATLAB code
Creating a Cloud-Based People Counter Using MATLAB
This demo uses a USB webcam and Computer Vision System Toolbox™ to detect faces within a video feed. The ThingSpeak™ Internet of Things (IoT) platform is used to track the number of people by transferring data to the cloud.
Visit this demo to see and discuss how MATLAB is used to:
- Perform real-time object recognition with live video acquisition
- Track features in a video and count the number of occurrences
- Create a web-based visualization updated with live data
A Fast Approach to Deep Learning with MATLAB
Deep learning models can achieve state-of-the-art accuracy in object classification, sometimes exceeding human-level performance. A tremendous amount of labeled data is required to train models, but transfer learning is a convenient way of applying deep learning without the need for a huge dataset and very long training time. In this demo, we show how MATLAB makes it easy for engineers to learn and apply transfer learning to their problems.
Building Predictive Maintenance Algorithms Using Physical Models
More companies are becoming aware of the importance of predictive maintenance of their equipment: the ability to reduce unexpected failures and intelligently schedule maintenance based on live machine health data. However, companies are challenged to collect the vast amounts of historical machine data required to build the predictive maintenance algorithm, including enough faulty data to create an accurate model. This model should not only warn of an imminent system failure, but also indicate which component has to be replaced.
But what if your organization doesn’t possess these data? Visit this demo to see how you can address the challenges of predictive maintenance by using MATLAB and Simulink® to:
- Build a physical model to generate additional machine data
- Use the model to simulate faulty conditions
- Use the data to train a predictive machine learning model
Simulating Large-Scale Electrical Networks in Real Time
Today’s power grids are among the most complex technological infrastructures ever built. The emerging smart grid and the increasing adoption of renewable energy systems raise new challenges for utilities and equipment suppliers. In this demo, large network layouts from third-party tools are automatically imported in MATLAB and re-implemented in Simulink, addressing the demand for test automation and simulation. Automatic code generation enables the rapid deployment of power-grid models on multicore real-time machines for hardware-in-the-loop testing and verification.
Visit this demo to learn how MATLAB and Simulink work together to:
- Facilitate the importing of large datasets and extract relevant information
- Use this information to automatically build models of complex systems
- Deploy these models on real-time machines to investigate hardware and prototype integration
Model-Based Design for a Motor Controller for Xilinx Zynq SoCs
From model to FPGA system-on-chip (SoC), see the implementation of a motor regulation algorithm on a heterogeneous SoC made of an FPGA and an ARM® processor through IP core generation workflows and automatic HDL and C code generation. This demo is based on a Xilinx® ZedBoard™ using an Analog Devices® motor control FMC board, and it runs a field-oriented controller (FOC) for a permanent magnet synchronous machine (PMSM).
Visit this demo to see and discuss how MATLAB is used to:
- Reduce dependency on hardware with simulation
- Avoid implementation errors with C and HDL code generation
- Perform faster hardware iterations with deployment automation
Virtual Commissioning and Verification of a Manufacturing Cell
Virtual commissioning and early design verification are key processes in today’s automation industry. Explore a model of a manufacturing cell to learn how you can combine your existing CAD models with mechatronic actuators to confidently make component sizing decisions. A robot arm with five degrees of freedom transfers parts between two conveyor belts. The model is used to determine mechanical loads, size motors, and determine overall power requirements. Optimization algorithms are used to determine optimal trajectories that minimize power consumption. Model-Based Design does not stop at pure simulation: See how you can use the model to produce fully verified supervisory algorithms ready to be deployed through automatic code generation in embedded environments.
Visit this demo to learn how to:
- Increase productivity by importing and reusing CAD models in Simulink
- Quickly model and integrate complex multidomain physical systems
- Reduce time-to-market by using simulation and automatic code generation
Design and Prototype a Digital Motion Control System
Position control of motion systems faces rising demands on throughput and accuracy that force machine designers to go to the limits of physical (mechanical) capabilities. A common limitation is hit when increasing machine performance is restricted by the flexibility of the connection between actuator and load. This problem is addressed by considering an experimental motion system that has purposefully been designed to show this behavior: If we can design a controller to deal with this for the experimental setup, we can use the same approach for a real industrial motion system.
The demo shows the main steps of Model-Based Design, including modeling, control design, and rapid prototyping.
Polyspace Code Verification
With growing complexity in embedded software, software testing and reviews are more time-consuming and challenging than ever before. High-integrity applications must ensure that they work as intended under all circumstances because the consequences of failures are catastrophic. Software testing can only show the presence but never the absence of errors. Formal methods–based, static analysis can exhaustively prove the absence of certain run-time errors and properties of generated or handwritten C and C++ code.
This demo shows how to apply Polyspace® static analysis to:
- Comply with MISRA® coding guidelines
- Achieve credits to IEC 61508, ISO 26262, IEC 62304, and DO-178 requirements
- Uncover security vulnerabilities according to CERT-C and ISO 17961
- Fix programming defects as soon as they are introduced
MathWorks in Education and Academia
MathWorks is deeply rooted in education and research with more than 5,000 universities around the world using MATLAB and Simulink software for engineering and math courses and lab work. MATLAB and Simulink are frequently used in many compelling student assignments and competitions including RoboCup and the Solar Challenge. Students and professors like to work with MATLAB because it not only gives them the opportunity to work on more interesting real-life problems and examples, but also provides a running start for graduates at the many technological companies that are using the software as part of their development process.
Learn about the activities MathWorks supports at universities, including project-based learning, MATLAB Online, MATLAB Mobile, and MathWorks Academic Online Training Suite.
MathWorks Training Services
Discover how courses from MathWorks Training Services can accelerate your use of MATLAB and Simulink. New and advanced users of all application areas can find training that suits their needs. Come by our booth to:
- Learn about training offerings—from public training throughout Benelux and Europe to online, self-paced training at a time and location that suits you best, we can tailor the training to your needs
- Explore a variety of training paths for users with specific job roles or application areas
- See training materials for various courses
- Interact with online, self-paced materials, including the free training material in MATLAB Academy™, which enables new users to get up to speed in only a few hours
MathWorks Consulting Services
MATLAB and Simulink are created to be used by people who are experts in their own field, not necessarily experts in technical computing or in leveraging software tools to optimize their Model-Based Design. MathWorks Consulting Services can help you get the most out of the software by jumpstarting your project or setting up best practices for improved design flows, freeing up your time for solving the problems you care about. Visit the booth to connect with one of our experienced consultants and ask them about their most successful engagements.